# LiteRT use cases

LiteRT is a set of tools that allows on-device machine learning. You can run your models on mobile, embedded, and edge devices. LiteRT use cases allow you to run use cases for image classification, object detection, image segmentation, and pose estimation.

Before you run the use cases, complete the preconditions mentioned in GStreamer command-line use cases.

- Image classification and display with LiteRT
The use cases use the Inceptionv3 LiteRT model to classify scenes from a single camera stream and either overlay or compose the classification labels.
- Image classification and encode with LiteRT
The use cases use the InceptionV3 LiteRT model to classify scenes from a single camera stream and either overlay or compose the classification labels, and then encode the stream.
- Audio classification decode and display with LiteRT
The use cases implement the YAMNet LiteRT model to classify and decode audio samples from a microphone and a file source.
- Object detection and display with LiteRT
The use cases use a YOLOX LiteRT model to identify the object in a scene. The use case is to either overlay or compose the bounding boxes over the detected objects, and then display the results.
- Object detection and encode with LiteRT
The use cases use a YOLOX LiteRT model to identify the object in a scene. The use case is to either overlay or compose the bounding boxes over the detected objects, and then encode this stream as an H.264 bitstream.
- Image segmentation and display with LiteRT
The use case implements the `deeplabv3_resnet50` LiteRT model to identify semantic segmentations in a scene from a camera stream. The use case is to compose the semantics and original video stream using qtivcomposer, and then display the results.
- Image segmentation and encode with LiteRT
The use case implements the `deeplabv3_resnet50` LiteRT model to compose the semantic segmentations and original video stream, encode this stream, and then multiplex it in an MP4 container.
- Pose estimation and display with LiteRT
The use cases implement the HRNet LiteRT model to process a single camera stream with pose estimation.
- Pose estimation and encode with LiteRT
The use cases implement the HRNet LiteRT model to process a single camera stream with pose estimation and encode the stream as an H.264 bitstream.
- Video super resolution and display with LiteRT
Video super resolution (VSR) is supported on Qualcomm AI Hub quantized INT8 models with 128 ×128 input resolution and 512 × 512 output resolution.
- Single stream from camera to RTSP with ML detection
Play a stream from the camera through RTSP on a media player (such as VLC).
- Four stream batching with LiteRT
This command line shows a four-stream batched AI inference, that is, an object detection use case from a video file.
- Object detection using USB camera source
The use case streams video from a USB webcam connected to the Qualcomm EVK. This webcam should be accessible as a `/dev/videoX` device. Additionally, you can perform object detection and preview the results.

Last Published: May 14, 2026

Previous Topic
 
Run machine learning use cases Next Topic

Image classification and display with LiteRT